От: Ron Johnson
Тема: Re: postgresql.conf
Дата: ,
Msg-id: 1059589963.7505.325.camel@haggis
(см: обсуждение, исходный текст)
Ответ на: postgresql.conf  ("scott.marlowe")
Список: pgsql-performance

Скрыть дерево обсуждения

postgresql.conf  ("scott.marlowe", )
 Re: postgresql.conf  (Josh Berkus, )
 Re: postgresql.conf  (Ron Johnson, )
 Re: postgresql.conf  ("Robert Treat", )
  Re: postgresql.conf  (Josh Berkus, )
   Re: postgresql.conf  (Ron, )
    Re: postgresql.conf  (Josh Berkus, )
     Re: postgresql.conf  ("scott.marlowe", )
      Re: postgresql.conf  (Josh Berkus, )
      Re: postgresql.conf  (cafweb, )

On Wed, 2003-07-30 at 11:59, scott.marlowe wrote:
> I'm looking at doing the example postgresql.conf files for the 7.4
> release. So far, the catagories we have would be a matrix of:
>
> -------------- Large Machine -- Small Machine
> Webserver
> OLAP
> OLTP
> Workstation
>
> But likely only one entry for workstation.

How about "General Purpose", for DBs that don't fit into any one
category?

> anyone have any advice on what they use in which situations and what we
> should include in the examples?
>
> I'm guessing OLTP needs things like FSM cranked up,
> OLAP (a for analytical) needs more shared buffers and sort memory
> Webserver might be better served just slightly higher values than default
> but well under those of either OLTP or OLAP...



--
+-----------------------------------------------------------------+
| Ron Johnson, Jr.        Home:              |
| Jefferson, LA  USA                                              |
|                                                                 |
| "I'm not a vegetarian because I love animals, I'm a vegetarian  |
|  because I hate vegetables!"                                    |
|    unknown                                                      |
+-----------------------------------------------------------------+




В списке pgsql-performance по дате сообщения:

От: Bruce Momjian
Дата:
Сообщение: Re: Mapping a database completly into Memory
От: Tom Lane
Дата:
Сообщение: Re: Why performance improvement on converting subselect